Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Documentation for Goal conditioning #5149

Merged
merged 12 commits into from
Mar 29, 2021
Merged

Documentation for Goal conditioning #5149

merged 12 commits into from
Mar 29, 2021

Conversation

vincentpierre
Copy link
Contributor

Proposed change(s)

Documentation for Goal conditioning

Useful links (Github issues, JIRA tickets, ML-Agents forum threads etc.)

Types of change(s)

  • Bug fix
  • New feature
  • Code refactor
  • Breaking change
  • Documentation update
  • Other (please describe)

Checklist

  • Added tests that prove my fix is effective or that my feature works
  • Updated the changelog (if applicable)
  • Updated the documentation (if applicable)
  • Updated the migration guide (if applicable)

Other comments

@vincentpierre vincentpierre self-assigned this Mar 17, 2021
@vincentpierre vincentpierre marked this pull request as ready for review March 18, 2021 19:23
awjuliani
awjuliani reviewed Mar 18, 2021
View reviewed changes
docs/Learning-Environment-Design-Agents.md Outdated

It is possible for agents to collect observations that will be treated as "goal".
A goal is used to condition the policy of the Agent, meaning that if the goal
changes, the behavior of the Agent will change as well. Note that this is true
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we should say that goals change the policy, which is a mapping from (observation -> action)

awjuliani
awjuliani reviewed Mar 18, 2021
View reviewed changes
Copy link
Contributor

@awjuliani awjuliani left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we want to refer to it as a "goal signal" instead of a goal?

awjuliani
awjuliani reviewed Mar 18, 2021
View reviewed changes
docs/Learning-Environment-Design-Agents.md Outdated
@@ -560,6 +562,36 @@ between -1 and 1.
of an entity to the `BufferSensor`.
- Normalize the entities observations before feeding them into the `BufferSensor`.

### Goal Signals

It is possible for agents to collect observations that will be treated as "goal."
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
It is possible for agents to collect observations that will be treated as "goal."
It is possible for agents to collect observations that will be treated as a "goal."

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think within this paragraph we should also call these goal signals, and not goals.

@vincentpierre
Update docs/Learning-Environment-Design-Agents.md
3e2a3e0 
Co-authored-by: Ervin T. <ervin@unity3d.com>
@vincentpierre vincentpierre merged commit 96a5aaf into main Mar 29, 2021
@delete-merged-branch delete-merged-branch bot deleted the goal-conditioning-docs branch March 29, 2021 18:58
@github-actions github-actions bot locked as resolved and limited conversation to collaborators Mar 29, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers

@ervteng ervteng ervteng approved these changes

@awjuliani awjuliani awjuliani approved these changes

Assignees

@vincentpierre vincentpierre

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@vincentpierre @ervteng @awjuliani